Impact of COVID-19 on Peer Support Specialists in the United States: Findings From a Cross-Sectional Online Survey

Peer support specialists (PSS) are a growing part of the mental health services workforce; they are employed in diverse settings and perform a broad range of roles (1). The Substance Abuse and Mental Health Services Administration defines a peer worker as a person “who uses his or her lived experience of recovery from mental illness and/or substance use disorder, plus skills learned in formal training … to deliver services in behavioral health settings to promote mind-body recovery and resilience” (2). In 2007, the Centers for Medicare and Medicaid Services deemed peer support an evidence-based practice and an important component of a state’s mental health system (3).

Although some PSS work as volunteers, peer support is now an organized paid workforce providing services in a range of service settings. The PSS workforce has grown substantially, from approximately 10,000 certified PSS in 2005 (4) to more than 25,000 in 2016 in the United States alone (5). As of 2020, 47 states and the District of Columbia offer peer specialist certification (6). The Veterans Health Administration currently employs several hundred peer support staff (7), and many mental health service organizations have adopted a recovery model that includes hiring PSS (8). Although much of the available research assessing peer support has methodological shortcomings (9, 10), there is some evidence that PSS improve outcomes for people with mental health issues. Specifically, PSS increase engagement and activation in other mental health services (11, 12), reduce the number of hospitalizations (10, 13, 14), improve self-efficacy (10, 15), and support recovery-related outcomes, such as hope and empowerment (14).

The COVID-19 pandemic has had widespread implications for physical, mental, and economic health and well-being. Nonpharmaceutical interventions, such as quarantine, have led to physical isolation, psychological distress, and misuse of substances (16, 17). Job loss was enormous both in the general workforce and among health care and social service workers specifically (18). Frontline workers have experienced increased workload and trauma, making them susceptible to burnout, depression, and posttraumatic stress disorder (17, 19, 20).

The pandemic also ushered in new, abrupt changes to the provision of mental health care. Mental health providers began to engage in new tasks as the pandemic persisted (21, 22). Telehealth is one such change. A 154% increase in telehealth visits during the last week of March 2020, compared with the same period in 2019 (23), reflects pandemic-related telehealth policy changes and a rapid transition to remote engagement of program participants with use of telephonic and virtual platforms. Accumulating information suggests broad satisfaction with telepsychiatry, higher rates of visit completion, and interest in its continuation postpandemic (24, 25). Additionally, mental health care providers, as a primary point of contact with the health care system for many vulnerable individuals, were rapidly required to learn to identify signs of COVID-related illness and mitigation strategies (26). In light of these and other significant changes brought on by the pandemic, this study aimed to assess its impact on work status, changes to job tasks, and new challenges among PSS based in the United States.

Methods

The authors formed a partnership with the National Association of Peer Supporters (NAPS) to conduct this cross-sectional, online survey. The Boston University Institutional Review Board approved all research procedures and materials. The researchers and NAPS worked collaboratively to develop the research questions and the methods, including five screening questions to ensure that individuals met inclusion criteria. The final survey contained 49 items as follows: demographic information, 12 items; work history and status, 17 items; changes to work tasks, 11 items; new challenges in work, 1 item; questions about technology, 7 items; and an open-ended comment question. All survey items were designed specifically for this study. We used a combination of item and response types (dichotomous and Likert response scales). We developed the items and procedures for survey implementation following the principles of the total design method (27) as well as participatory research methods (28). Participatory action research (PAR) is a process for engaging key stakeholders that is designed to be empowering for end-users and that enables greater control over circumstances—in this case the delivery of peer support services (28). In keeping with PAR, PSS were involved in every aspect of this study, from conceptualization to interpretation of the data to preparing the manuscript for publication.

We undertook several rounds of item refinement, followed by multiple iterations of pilot testing in Qualtrics. In addition, two colleagues unfamiliar with the study purpose tested the online survey several times to check for clarity and skip logic. If an error was detected, several authors reviewed the survey items again.

Survey Participants

The final version of the online survey was distributed through the NAPS Listserv, composed of approximately 6,000 individuals. NAPS sent four e-mails, first to invite individuals to complete the survey and three subsequent e-mails to prompt nonresponders. E-mail recipients could forward the survey link to colleagues or organizations meeting eligibility criteria, allowing for snowball sampling. Individuals were eligible for the survey if they lived in the United States, were 18 years or older, and had worked or volunteered as a PSS for at least 5 hours a week in February 2020, prior to the U.S. COVID-19 crisis. Because of regional and organizational differences in peer job titles (such as peer coaches, peer advocates, recovery specialists, and recovery coaches), we allowed all peer workers to participate. Inclusion criteria were verified by using the screening questions. Individuals not meeting all inclusion criteria were thanked for their interest and told that they were not eligible to proceed.

Data Collection and Monitoring

Data were collected from May 18 through June 22, 2020. Data collection was monitored closely to ensure that surveys were being accessed and completed without difficulty. A total of 1,772 responses were recorded. We eliminated duplicates (N=85), ineligible responses (i.e., those who did not meet all eligibility criteria; N=223), and those who declined or withdrew (i.e., those who began the survey but did not proceed past consent; N=184). The final sample size for analysis was 1,280.

Data Analysis

Responses were downloaded from Qualtrics into SPSS, version 25.0 (29) for cleaning and analysis. We checked for out-of-range values, valid responses, and difficulties with skip logic. To determine differences in before-to-after COVID-19 responses, we examined frequency tables; conducted cross-tabulations, chi-square tests, paired t tests, and repeated-measures analysis of variance. In keeping with our PAR approach, NAPS was involved in advising us about data analysis and in interpretation of the quantitative portion of the survey.

Participants who responded yes to the question “Since the coronavirus began, have you engaged in any new tasks in your peer specialist role?” were asked to describe new work tasks as an open-ended response. Responses were analyzed with content analysis (30) by two authors (WEA, LM). Codes were developed inductively and revised throughout the analytic process. Similar codes were grouped together to form broader categories that emerged from the data. We did not use preidentified codes or categories because of a lack of existing data on the subject. The lead author (WEA) reviewed open-ended responses multiple times to identify initial emerging themes. The two aforementioned authors assigned codes to participant responses, and both reviewed final codes and categorization until reaching consensus.

Results

Most respondents were female (69%) and White (73%), and ages ranged from 18 to 89 (mean=50) (Table 1). More than half had an associate’s, bachelor’s, graduate, or professional degree. Respondents resided in all 50 states and provided services in large cities, small towns, and suburban and rural areas. Forty-one percent of respondents reported currently receiving public benefits, and 39% earned less than $30,000 in 2019. Respondents were employed in a variety of settings. Among those who responded to the survey item, 66% (N=803) reported working full-time before the pandemic.

Work Status and Changes to Employment

A total of 109 respondents (9%) reported that they were laid off or lost a job because of COVID-19 (Table 2). Among all respondents, 215 (17%) reported that a peer colleague was similarly affected. Respondents’ job loss was unrelated to geographic region, type of community in which they worked, or organizational type (data not shown). Among the 109 respondents who lost a peer position since the beginning of the pandemic, 71 (65%) had been employed in their role for 2 years or longer. We examined demographic characteristics (age, gender, race, ethnicity, and level of education) and found that none was significantly associated with job loss (data not shown). Six percent of respondents who had been employed prepandemic reported having no paid or volunteer PSS position at the time of survey completion in May or June 2020 (Table 3). The shifts we observed in employment situations from prepandemic to the time of survey completion were statistically significant (χ²=867.40, df=6, p<0.001). Of the 1,209 respondents for whom we had complete data, 1,085 (90%) reported working full- or part-time prepandemic, and 1,048 (87%) reported working at the time they completed the survey. Fewer individuals reported being engaged in any employment situation—whether that was full- or part-time work, volunteering, or working and volunteering—while at the same time, 75 individuals moved to the neither working nor volunteering at the time of survey completion.

The number of PSS reporting an inability to pay their monthly bills increased from 287 (24%) before the pandemic to 385 (33%) at the time of survey completion. Ninety-four respondents (7%) reported losing some or all benefits because of COVID-19, and 171 (13%) reported reduced earnings. Almost two-thirds of respondents (N=742, 65%) reported that the pandemic affected their personal financial situation less than it affected the financial situation of most other people. (These percentages are based on the number of responses to the survey items, which varied.)

Changes to Job Tasks

Respondents reported changes to their job tasks since the COVID-19 pandemic began (Tables 3 and 4). First, they reported decreases in administrative tasks, provision of individual support, training, group facilitation, and systems advocacy. Second, since the coronavirus began, approximately three-quarters (N=761, 73%) of those who continued to work reported engaging in new tasks in five main categories: use of technology, connecting individuals to resources, COVID-19–specific tasks, tasks not typically engaged in by peer staff, and tasks related to provision of support. The most commonly reported new tasks were related to technology, such as providing support remotely, learning new technology, and facilitating online groups. Respondents reported increased resource coordination (e.g., connecting individuals whom they supported to food, housing, and transportation) and tasks related to COVID-19 (e.g., getting personal protective equipment, performing temperature screenings, and conducting training). Tasks not typically engaged in by peer staff that increased included case management and administration, and peer-centered tasks included increased outreach and support of their colleagues. Table 4 provides additional information and sample quotations from the survey.

Finally, among the 1,086 PSS who reported current employment and who answered a question regarding service user changes, some (N=487, 45%) reported that the number of individuals they supported increased during the pandemic, whereas others (N=316, 29%) reported that the number decreased. More than half (N=621, 57%) reported that more individuals were reaching out for support since the pandemic began.

Challenges and Support

We provided a checklist of possible adverse experiences as a result of the pandemic (Table 5). Regarding the individuals they supported, almost all PSS respondents (92%) indicated that they were experiencing increased isolation, and more than half reported that they were facing challenges related to increased symptoms affecting mental health, increased substance use, food shortages, and the possibility of job loss. We also asked PSS whether they themselves were experiencing additional challenges related to their PSS roles since the pandemic began. Most reported experiencing isolation and communication challenges.

On a Likert scale (1, very satisfied; 5, very unsatisfied), respondents expressed being satisfied with the amount of support they were receiving since the pandemic began, including a high level of satisfaction with their organization (mean±SD rating=2.18±1.27), their supervisor (1.99±1.23), and their coworkers (1.91±1.09). When asked about specific types of support (social-emotional, training-educational, or material), mean responses ranged from 2.19 to 2.32.

Almost three-quarters of the 1,070 respondents who reported being employed at the time of the survey and who answered the question regarding benefits that may have accrued since the pandemic began reported experiencing positive impacts of their role as PSS (N=785, 73%). Benefits reported included less time spent commuting, the ease of working from home, learning new technological skills, and being able to support individuals who had been unable or unwilling to receive services because of transportation challenges or preference.

Discussion

A total of 1,280 U.S.-based PSS qualified for, consented to, and responded to a cross-sectional, online survey conducted in May and June 2020 about the impact of COVID-19 on various aspects of their work. To meet inclusion criteria, respondents had to have been employed or volunteering at least 5 hours per week in February 2020, prior to the pandemic. Approximately 9% of respondents reported that they lost a job or were laid off because of COVID-19. Although this is a substantial proportion, it is far lower than the 15% of U.S. adults reporting COVID-19–related job loss (31). Furthermore, in the early months of the pandemic, most respondents believed that the pandemic affected their personal financial situation less than it affected those of most others.

Our results suggest that the work of PSS changed to adapt to the pandemic, with most respondents engaged in new job tasks, including using technology and connecting the persons they served with community resources. Conversely, respondents reported engaging less frequently in many standard tasks, particularly in terms of providing individual support and facilitating groups. These changing roles and tasks have implications for hiring, training, supervising, and supporting PSS. Importantly, additional technological training and equipment are needed, because PSS have been required to provide not only telehealth and remote group support but also technological support to the individuals they serve. These findings comport with those of an early review of the effect of the pandemic on mental health services, which found that individuals benefited from both the practical and the emotional support offered by peers (32). Creative and proactive deployment of digital and online support facilitated both one-to-one connections and group sessions as well as recreational and social activities (32). Although we can identify no other studies assessing job task changes among PSS as a result of the pandemic, our findings align with recent work showing community health workers engaging in multiple new roles and tasks during recent pandemics, including COVID-19 (33).

Respondents reported that the individuals they supported were experiencing significant additional challenges, ranging from material to symptomatic to social challenges and including substantial isolation, substance use issues, housing instability, food shortages, and interpersonal and family violence. This finding is consistent with those of previous studies related to major concerns among individuals living with mental illness, including during the pandemic (26, 34–36). Most respondents reported that more individuals were reaching out for peer support services since the pandemic began, perhaps adding to the demand for PSS services.

Respondents also reported that they themselves were experiencing pandemic-related challenges, especially isolation and difficulties with communication. Despite these challenges, most respondents reported feeling satisfied with the amount of support their organizations, supervisors, and coworkers were providing, which suggests that agencies were able to adequately support PSS during the first few months of the pandemic.

Notably, almost three-quarters of respondents reported positive impacts resulting from the pandemic, many related to the adoption of technologies that allowed for the provision of remote support. The need to address the use of telehealth and technological advances to provide remote support is garnering more attention. For example, Fortuna and colleagues (37) developed and tested a digital peer support intervention and found it to be a promising way to increase the capacity of PSS to use specific digital technology features in their work.

Guinart and colleagues (24) reported that most mental health care providers surveyed in April and May 2020 expressed positive attitudes toward telepsychiatry and expressed interest in its continuity after the pandemic has resolved. However, it will remain critical for the PSS workforce to address the challenges of technological literacy and lack of access to technology that other mental health care providers report.

Moreno and colleagues (22) argued that the mental health care system must adopt greater inclusion of service users as it adapts to the pandemic and implements long-lasting changes. This inclusion may need to go beyond supporting PSS involvement to focusing attention on larger implementation issues (38). Although the findings are based on PSS self-report, they suggest success within the peer support workforce in adapting to new challenges related to COVID-19—success on which to capitalize.

Several limitations of this study should be noted. First is the representativeness of this sample in terms of the entire peer support workforce. However, given the similarity of the demographic characteristics of this survey’s respondents to those of samples in previous studies of PSS (31, 39, 40), we are reasonably confident that our respondents are representative of the larger cadre of specialists.

We believe that we have adequately represented PSS of color given that the demographic characteristics of survey respondents appear to be in line with other data that have been captured (39). Although our survey represented views of PSS of color, we were not able to adequately identify or explore the potentially disproportionate impact that the pandemic has had on people from many marginalized and underrepresented groups within the PSS and support-recipient populations. This is critical because the pandemic has been shown to shed light on and exacerbate previously existing health inequalities (22).

As with any survey, we relied to some degree on the accurate interpretation of items and respondents’ recollection. The survey was carefully constructed with participation from PSS who are familiar with this workforce by using a participatory research approach. Items were vetted and refined numerous times for clarity. However, the effects of measurement error cannot be discounted, because all items were developed for the purposes of this survey.

In addition to these limitations, our survey was implemented in a short, specific time frame (May and June 2020), leaving open the possibility of a time-cohort effect. Different responses might have been found during a different time frame and with a different group of respondents. Finally, it is possible that PSS without easy access to the Internet may not be represented in these results. However, surveyed PSS suggest that the pandemic forced them to rely more than ever on technology and the Internet for their job tasks, suggesting that a substantial proportion of PSS have access to e-mail and the Internet.

Conclusions

Results suggest that PSS are filling the needs that individuals served in the mental health system have in their daily lives—needs that have increased or become more visible since the pandemic began. These may include needs that mental health programs and clinical staff considered outside their purview or regular job duties. PSS reported using creative and innovative approaches to addressing the increased isolation and other adverse effects of the pandemic and closure of many programs and services. Similar findings have been reported by other researchers (22, 32). It appears that PSS and their organizations demonstrated the flexibility to address these unplanned needs and to continue providing critical support services during the pandemic. Implications include the need to know more about how the mental health workforce and PSS are being affected during the pandemic and beyond. The reported unexpected benefits arising from the pandemic should be examined to ensure the continuation of flexible, creative, and responsive mental health services.